2014 · 33 citations · 25 references
ComputingComputer ScienceComputer NetworksFault ToleranceFault-tolerant SystemsNetwork ArchitectureNetwork Fault ToleranceRoutingRouting AlgorithmsRouting ProtocolsSoftware-defined NetworkingFault-tolerant NetworkingFault-tolerant NetworksFault-tolerant RoutingFault-tolerant Software-defined NetworkingNode FailuresOpenflowResilient NetworkingScalability
One of the main concerns on SDN is relative to its ability to quickly react to network failures, while limiting both the control-plane overhead and the additional forwarding state kept by data-plane devices. Despite its practical importance, this concern is often overlooked in OpenFlow-based proposals. In this paper, we propose a new architecture, called IBSDN, in which a distributed routing protocol flanks OpenFlow to improve network robustness, reaction to failures, and controller scalability. In deeply exploring this idea, we complement our architecture with data-plane triggered mechanisms that improve its efficiency. We prove that the resulting solution ensures robustness for any combination of topological failures, and quickly reduces the path stretch. Finally, experimenting with a prototype implementation, we show that our approach is practical and overcomes the main limitations of previous work.
25
Measuring ISP topologies with rocketfuel
Neil Spring, Ratul Mahajan, David Wetherall · 2002 · 1.3K citations
HyperFlow: a distributed control plane for OpenFlow
Amin Tootoonchian, Yashar Ganjali · 2010 · 902 citations
SIMPLE-fying middlebox policy enforcement using SDN
Zafar Ayyub Qazi, Cheng-Chun Tu, Luis Chiang et al. · 2013 · 648 citations · Full text
Abstractions for network update
Mark Reitblatt, Nate Foster, Jennifer Rexford et al. · 2012 · 580 citations · Full text
Extending Networking into the Virtualization Layer.
Ben Pfaff, Justin Pettit, Keith Amidon et al. · 2009 · 503 citations